Safety & Trust
Protection is the architecture, not a policy page
Aurelia was designed child-safe from the first database table upwards. This page explains, in plain terms, how that protection actually works — not just what we promise.
The foundations
Four decisions that shape everything else
These are structural choices, not optional settings. They apply to every account, every piece of work and every interaction on the platform.
Invitation-only onboarding
There is no open self-registration for children. Every child joins through a verified parent/guardian or school invitation, so we know who is bringing a child onto the platform before an account exists.
Guardian approval
A guardian must approve a child's account and must approve before any piece of work leaves the family and becomes visible to a class, club or wider audience.
Database-enforced permissions
Who can see what is not a setting in the interface that a bug could bypass. Permissions are enforced with row-level security in the database itself, so access control holds even if a screen is built incorrectly.
Separated age worlds
The under-16 environment and the 16+ alumni environment are separate by design. There is no shared social surface, messaging channel or discovery feed connecting the two.
Moderation
Review by people, not just filters
Automated tools help us spot potential issues quickly, but the final decisions about what is shared and how concerns are handled sit with trained people who carry that responsibility.
- Work intended for wider sharing passes through review before it becomes visible beyond the family or class.
- Trained staff, not automated filters alone, make the final call on borderline or reported content.
- Clubs and challenges carry named adult supervision rather than open, unmoderated interaction.
- Every account has clear, accessible routes to report a concern about content or conduct.
Accountability
How we stay honest about what happens
Trust is easier to claim than to demonstrate. These are the mechanisms that let decisions be checked, questioned and improved over time.
Audit trails
Sensitive actions — approvals, publications, permission changes, access to a child's data — are written to an auditable trail so decisions can be reviewed after the fact, not just trusted in the moment.
Incident response
Safeguarding and security concerns follow a defined path: triage, investigation, containment and, where appropriate, escalation to guardians, schools or the relevant authorities.
Human review
Automated checks flag possible issues, but decisions that affect a child's account, work or safety are made or confirmed by a person with safeguarding responsibility.
Bounded AI
AI assistance is age-banded, limited in scope and always labelled, so it supports a young person's thinking without acting unsupervised on their behalf or their data.
Detail matters here, because a policy page is not enough
This page is deliberately thorough. The reassurance a family or school needs is not a soft promise — it is a clear account of the specific mechanisms holding a child's safety in place, so anyone can see exactly how the protection works.
Read more
Related pages
Safety touches every part of Aurelia. These pages go into more detail on specific areas.
Bring Aurelia to your family or school
Aurelia is invitation-only. Tell us who you are and we will guide you through verified onboarding.